Abstract
This paper further investigates results from Adams and Wu (J Appl Meas, 10(4), 355–370, 2009) who developed a User Defined Fit Statistic to test the fit of linear combinations of items (or parameters) using linear combinations of individual contributions to the sufficient statistics and comparing them to their expectations. The User Defined Fit Statistic has the ability to identify violations of local independence and/or violations of uni dimensionality using a priori knowledge of the data. In a lot of cases, individual item (or parameter) residual based fit statistics are unable to identify such model violations. Simulations showed the User Defined Fit Statics’ ability to identify item response model violations, when local item dependence and multi dimensionality was present in simulated data. Two real datasets were also analysed, where local item dependence and multi dimensionality were again identified, displaying how useful these fit statistics can be (with a priori knowledge of the data) in identifying such model violations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Adams, R. J., & Wilson, M. R. (1996). Formulating the Rasch model as a mixed coefficient multinomial logit. In G. Engelhard Jr. & M. Wilson (Eds.), Objective measurement: Theory into practice (Vol. 3, pp. 143–166). Norwood: Ablex.
Adams, R. J., & Wu, M. L. (2009). The construction and implementation of user-defined fit tests for use with marginal maximum likelihood estimation and generalised item response models. Journal of Applied Measurement, 10(4), 355–370.
Adams, R. J., Wilson, M. R., & Wu, M. L. (1997). Multilevel item response models: An approach to errors in variables regression. Journal of Educational and Behavioural Statistics, 22(1), 47–76.
Andrich, D. A., & Marais, I. (2005). Studies on the effect of violations of local independence on scale in Rasch models. Funded by an ARC grant with industry partners MCEETYA, IIEP, and ACER.
Goldstein, H. (1980). Dimensionality, bias, independence and measurement scale problems in latent trait test score models. The British Journal of Mathematical and Statistical Psychology, 33, 234–246.
Lord, F. M., & Novick, M. R. (1968). Statistical theories of mental test scores. Reading: Addison-Wesley.
Wright, B. D., & Masters, G. N. (1982). Rating scale analysis: Rasch measurement. Chicago: MESE Press.
Wu, M. L. (1997). The development and application of a fit test for use with marginal maximum likelihood estimation and generalised item response models. Unpublished Master of Education Thesis, Melbourne University.
Wu, M. L., Adams, R. J., Wilson, M. R., & Haldane, S. (2007). ACER ConQuest version 2.0 [computer program]. Camberwell: ACER Press, Australian Council for Educational Research.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Urbach, D. (2013). Further Implementation of User Defined Fit Statistics. In: Zhang, Q., Yang, H. (eds) Pacific Rim Objective Measurement Symposium (PROMS) 2012 Conference Proceeding. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-37592-7_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-37592-7_5
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-37591-0
Online ISBN: 978-3-642-37592-7
eBook Packages: Humanities, Social Sciences and LawEducation (R0)